{"id":"https://openalex.org/W4410341150","doi":"https://doi.org/10.1109/tcasai.2025.3569511","title":"A Configurable Floating-Point Fused Multiply-Add Design With Mixed Precision for AI Accelerators","display_name":"A Configurable Floating-Point Fused Multiply-Add Design With Mixed Precision for AI Accelerators","publication_year":2025,"publication_date":"2025-05-13","ids":{"openalex":"https://openalex.org/W4410341150","doi":"https://doi.org/10.1109/tcasai.2025.3569511"},"language":"en","primary_location":{"id":"doi:10.1109/tcasai.2025.3569511","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcasai.2025.3569511","pdf_url":null,"source":{"id":"https://openalex.org/S4404675360","display_name":"IEEE transactions on circuits and systems for artificial intelligence.","issn_l":"2996-6647","issn":["2996-6647"],"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Circuits and Systems for Artificial Intelligence","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5064837871","display_name":"Farzad Niknia","orcid":"https://orcid.org/0000-0002-4062-3638"},"institutions":[{"id":"https://openalex.org/I12912129","display_name":"Northeastern University","ror":"https://ror.org/04t5xt781","country_code":"US","type":"education","lineage":["https://openalex.org/I12912129"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Farzad Niknia","raw_affiliation_strings":["Department of Electrical and Computer Engineering, Northeastern University, Boston, MA, USA","Department of Electrical and Computer Engineering, Northeastern University Boston, MA, USA"],"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, Northeastern University, Boston, MA, USA","institution_ids":["https://openalex.org/I12912129"]},{"raw_affiliation_string":"Department of Electrical and Computer Engineering, Northeastern University Boston, MA, USA","institution_ids":["https://openalex.org/I12912129"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100717659","display_name":"Ziheng Wang","orcid":"https://orcid.org/0000-0001-9668-7318"},"institutions":[{"id":"https://openalex.org/I12912129","display_name":"Northeastern University","ror":"https://ror.org/04t5xt781","country_code":"US","type":"education","lineage":["https://openalex.org/I12912129"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ziheng Wang","raw_affiliation_strings":["Department of Electrical and Computer Engineering, Northeastern University, Boston, MA, USA","Department of Electrical and Computer Engineering, Northeastern University Boston, MA, USA"],"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, Northeastern University, Boston, MA, USA","institution_ids":["https://openalex.org/I12912129"]},{"raw_affiliation_string":"Department of Electrical and Computer Engineering, Northeastern University Boston, MA, USA","institution_ids":["https://openalex.org/I12912129"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100417119","display_name":"Shanshan Liu","orcid":"https://orcid.org/0000-0001-6226-2880"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shanshan Liu","raw_affiliation_strings":["School of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu, China"],"affiliations":[{"raw_affiliation_string":"School of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu, China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5080322790","display_name":"Pedro Reviriego","orcid":"https://orcid.org/0000-0003-2540-5234"},"institutions":[{"id":"https://openalex.org/I88060688","display_name":"Universidad Polit\u00e9cnica de Madrid","ror":"https://ror.org/03n6nwv02","country_code":"ES","type":"education","lineage":["https://openalex.org/I88060688"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Pedro Reviriego","raw_affiliation_strings":["Departamento de Ingeniería de Sistemas Telemáticos, Escuela Técnica Superior de Ingeniería de Telecomunicación, Universidad Politécnica de Madrid, Madrid, Spain","Departamento de Ingenier\u00eda de Sistemas Telem\u00e1ticos, Escuela T\u00e9cnica Superior de Ingenier\u00eda de Telecomunicaci\u00f3n,, Universidad Polit\u00e9cnica de Madrid, Madrid, Spain"],"affiliations":[{"raw_affiliation_string":"Departamento de Ingeniería de Sistemas Telemáticos, Escuela Técnica Superior de Ingeniería de Telecomunicación, Universidad Politécnica de Madrid, Madrid, Spain","institution_ids":[]},{"raw_affiliation_string":"Departamento de Ingenier\u00eda de Sistemas Telem\u00e1ticos, Escuela T\u00e9cnica Superior de Ingenier\u00eda de Telecomunicaci\u00f3n,, Universidad Polit\u00e9cnica de Madrid, Madrid, Spain","institution_ids":["https://openalex.org/I88060688"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054396881","display_name":"Zhen Gao","orcid":"https://orcid.org/0000-0002-2709-0216"},"institutions":[{"id":"https://openalex.org/I162868743","display_name":"Tianjin University","ror":"https://ror.org/012tb2g32","country_code":"CN","type":"education","lineage":["https://openalex.org/I162868743"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhen Gao","raw_affiliation_strings":["School of Electrical and Information Engineering, Tianjin University, Tianjin, China","School of Information and Engineering, Tianjin University, Tianjin, China"],"affiliations":[{"raw_affiliation_string":"School of Electrical and Information Engineering, Tianjin University, Tianjin, China","institution_ids":["https://openalex.org/I162868743"]},{"raw_affiliation_string":"School of Information and Engineering, Tianjin University, Tianjin, China","institution_ids":["https://openalex.org/I162868743"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5083219829","display_name":"Paolo Montuschi","orcid":"https://orcid.org/0000-0003-2563-2250"},"institutions":[{"id":"https://openalex.org/I177477856","display_name":"Polytechnic University of Turin","ror":"https://ror.org/00bgk9508","country_code":"IT","type":"education","lineage":["https://openalex.org/I177477856"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Paolo Montuschi","raw_affiliation_strings":["Dipartimento di Automatica e Informatica, Politecnico di Torino, Torina, Italy"],"affiliations":[{"raw_affiliation_string":"Dipartimento di Automatica e Informatica, Politecnico di Torino, Torina, Italy","institution_ids":["https://openalex.org/I177477856"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5001979328","display_name":"Fabrizio Lombardi","orcid":"https://orcid.org/0000-0003-3152-3245"},"institutions":[{"id":"https://openalex.org/I12912129","display_name":"Northeastern University","ror":"https://ror.org/04t5xt781","country_code":"US","type":"education","lineage":["https://openalex.org/I12912129"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Fabrizio Lombardi","raw_affiliation_strings":["Department of Electrical and Computer Engineering, Northeastern University, Boston, MA, USA","Department of Electrical and Computer Engineering, Northeastern University Boston, MA, USA"],"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, Northeastern University, Boston, MA, USA","institution_ids":["https://openalex.org/I12912129"]},{"raw_affiliation_string":"Department of Electrical and Computer Engineering, Northeastern University Boston, MA, USA","institution_ids":["https://openalex.org/I12912129"]}]}],"institutions":[],"countries_distinct_count":4,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5064837871"],"corresponding_institution_ids":["https://openalex.org/I12912129"],"apc_list":null,"apc_paid":null,"fwci":1.5687,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.83710837,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":97,"max":99},"biblio":{"volume":"2","issue":"3","first_page":"248","last_page":"261"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10848","display_name":"Advanced Multi-Objective Optimization Algorithms","score":0.9642999768257141,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10848","display_name":"Advanced Multi-Objective Optimization Algorithms","score":0.9642999768257141,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11697","display_name":"Numerical Methods and Algorithms","score":0.9426000118255615,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5238694548606873},{"id":"https://openalex.org/keywords/point","display_name":"Point (geometry)","score":0.49918127059936523},{"id":"https://openalex.org/keywords/floating-point","display_name":"Floating point","score":0.41138404607772827},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.34106194972991943},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.21459320187568665},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.18831786513328552}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5238694548606873},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.49918127059936523},{"id":"https://openalex.org/C84211073","wikidata":"https://www.wikidata.org/wiki/Q117879","display_name":"Floating point","level":2,"score":0.41138404607772827},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.34106194972991943},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.21459320187568665},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.18831786513328552},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tcasai.2025.3569511","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcasai.2025.3569511","pdf_url":null,"source":{"id":"https://openalex.org/S4404675360","display_name":"IEEE transactions on circuits and systems for artificial intelligence.","issn_l":"2996-6647","issn":["2996-6647"],"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Circuits and Systems for Artificial Intelligence","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","score":0.6800000071525574,"display_name":"Affordable and clean energy"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":40,"referenced_works":["https://openalex.org/W605824955","https://openalex.org/W1901616594","https://openalex.org/W2037346171","https://openalex.org/W2070273314","https://openalex.org/W2100974190","https://openalex.org/W2113893001","https://openalex.org/W2142464516","https://openalex.org/W2144921006","https://openalex.org/W2147479230","https://openalex.org/W2224617858","https://openalex.org/W2473178953","https://openalex.org/W2541452119","https://openalex.org/W2545221752","https://openalex.org/W2781383697","https://openalex.org/W2801795932","https://openalex.org/W2901312569","https://openalex.org/W2914973130","https://openalex.org/W2924943819","https://openalex.org/W2963287959","https://openalex.org/W2963367920","https://openalex.org/W2981849318","https://openalex.org/W2982144630","https://openalex.org/W3021029305","https://openalex.org/W3132455321","https://openalex.org/W3170647102","https://openalex.org/W3187235458","https://openalex.org/W3194056411","https://openalex.org/W3217237164","https://openalex.org/W4210298930","https://openalex.org/W4221036944","https://openalex.org/W4308083739","https://openalex.org/W4312872667","https://openalex.org/W4320030813","https://openalex.org/W4324142375","https://openalex.org/W4360831846","https://openalex.org/W4378800959","https://openalex.org/W4386446991","https://openalex.org/W4391326594","https://openalex.org/W4392011391","https://openalex.org/W4404520468"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"Hardware":[0],"accelerators":[1],"for":[2,14,97,278],"deep":[3],"learning":[4],"in":[5,75,90,128,135,256,271],"artificial":[6],"intelligence":[7],"applications":[8],"must":[9],"often":[10],"meet":[11],"stringent":[12],"constraints":[13],"accuracy":[15,198],"and":[16,59,71,94,100,134,175,228,243,251,261],"throughput.":[17],"In":[18,34,79],"addition":[19,80],"to":[20,81,117,166,192,216,223,267],"architecture/algorithm":[21],"improvements,":[22],"high":[23],"performance":[24],"computational":[25,197],"techniques":[26],"such":[27],"as":[28,57,61,63],"mixed":[29,126,179],"precision":[30,46,127,180,186,191,264],"are":[31,73,142,281],"also":[32,233],"required.":[33],"this":[35],"paper,":[36],"a":[37,248],"floating-point":[38],"(FP)":[39],"fused":[40],"multiply-add":[41],"(FMA)":[42],"unit":[43],"supporting":[44],"mixed/multiple":[45],"is":[47,88,131,232,254],"proposed.":[48,234],"A":[49],"wide":[50],"range":[51],"of":[52,108,151,164,213,258],"conventional":[53],"FP":[54,102],"formats":[55,65],"(such":[56],"half":[58],"single)":[60],"well":[62],"emerging":[64],"(including":[66],"E4M3,":[67],"E5M2,":[68],"DLFloat,":[69],"BFLoat16":[70],"TF32)":[72],"supported":[74,262],"the":[76,85,92,106,138,146,152,159,172,181,184,206,211,225,229,240,272],"proposed":[77,86,112,160],"design.":[78],"all":[82,201],"these":[83],"formats,":[84],"design":[87,161,253],"flexible":[89,252],"manipulating":[91],"exponent":[93],"mantissa":[95],"lengths":[96],"8,":[98],"16":[99],"32-bit":[101],"numbers":[103],"based":[104,144,170],"on":[105,145,171],"needs":[107],"an":[109],"application.":[110],"The":[111],"FMA":[113,122,157,182],"can":[114],"be":[115],"configured":[116],"support":[118],"either":[119],"multiple":[120,168],"normal":[121,156],"operations,":[123,158],"or":[124],"alternatively":[125],"ASIC.":[129],"It":[130,195],"fully":[132],"pipelined":[133],"each":[136],"cycle,":[137],"input":[139],"bit":[140],"streams":[141],"processed":[143],"provided":[147],"configuration,":[148],"so":[149],"independent":[150],"previous":[153],"cycles.":[154],"For":[155,178],"utilizes":[162],"sharing":[163],"resources":[165],"parallelize":[167],"operations":[169,215],"available":[173,241],"hardware":[174,242,259],"required":[176],"precision.":[177],"accumulates":[183],"lower":[185],"dot":[187,203,226],"products":[188,204,227],"into":[189],"higher":[190],"avoid":[193],"overflow/underflow.":[194],"improves":[196],"by":[199],"adding":[200],"possible":[202],"at":[205],"same":[207],"time":[208],"while":[209],"decreasing":[210],"number":[212],"rounding":[214,218],"prevent":[217],"errors.":[219],"An":[220],"innovative":[221],"method":[222],"accumulate":[224],"aligned":[230],"addend":[231],"By":[235],"considering":[236],"tradeoffs":[237],"between":[238],"reusing":[239],"removing":[244],"unnecessary":[245],"complex":[246],"units,":[247],"more":[249],"efficient":[250],"attained":[255],"terms":[257],"metrics":[260],"different":[263],"computation":[265],"compared":[266],"other":[268],"designs":[269],"found":[270],"technical":[273],"literature.":[274],"Extensive":[275],"simulation":[276],"results":[277],"comparative":[279],"analysis":[280],"provided.":[282]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-03-20T23:20:44.827607","created_date":"2025-10-10T00:00:00"}